There has been a substantial progress in clinical application of molecularly targeted proteinaceous cytotoxins for the treatment of glioblastoma multiforme (GBM). Several cytotoxins have moved relatively quickly, as for novel anti-cancer drugs, from bench to clinic and these studies brought invaluable translational information. Based on clinical trials and laboratory research, we hypothesize that there are three important factors linked to the overall cytotoxins' clinical utility. One is the target specificity. Another factor impacting efficacy of cytotoxins is the relative percentage of patients' population that can be predicted to be the responders to a cytotoxin. And the third factor, an efficient delivery of the cytotoxin directly to the tumor site using convection-enhanced delivery (CED) has been proven to be effective in many cases, but it is ordinarily not monitored. These hypotheses will be tackled experimentally, since we have all the tools needed to do so. With regard to target specificity, it was found that a vast majority of GBM patients over-express binding sites for interleukin 13 (IL-13), IL-13Ra2. Furthermore, another receptor, EphA2 tyrosine kinase receptor, was found to be elevated in a large number of GBM specimens and cell lines, but not in normal brain. A prototype cytotoxin targeting EphA2 has been successfully generated. With regard to the population of patients being potential responders to the cytotoxins, over-expression of IL-13Ra2 and Epha2 together is present in almost 100% of patients with GBM and thus it is an ideal situation in which all patients could be predictably responders or eligible for treatment. With regard to recombinant cytotoxins' delivery to GBM, it was demonstrated that using gadolinium and labeled human serum albumin (HSA) a faithful monitoring of drug distribution through CED can be obtained. In addition, some breeds of dogs express molecular fingerprints similarly to human GBM and represent an attractive, large animal pre-clinical model of human disease; the CED has already been applied to the central nervous system of dogs. Therefore, three Specific Aims are proposed. Specific Aim #1 is to optimize a novel molecularly targeted anti-GBM agent that exploits specific over-expression of EphA2 receptor in GBM. We will continue the work on an ephrinA1-based bacterial toxin-containing cytotoxin that exhibit potent and specific GBM cell killing. Specific Aim #2 is to explore combinatorial targeting of EphA2 and IL- 13Ra2 receptors with the cytotoxins. The combination therapy will be tested for an anti-tumor efficacy in in vitro and in vivo models of human GBM. In Specific Aim #3, Phase I clinical trial in dogs with malignant gliomas using combinatorial recombinant cytotoxins approach will be performed. The distribution of the CED- delivered drugs will be monitored long- and short-term by employing MRI and PET and using Gd-HSA and [68]Ga-HSA, respectively. The results of this project will permit further successful clinical application of recombinant cytotoxins, a highly promising class of anti-GBM drugs.